Compare AI models

Specifications
Parameters
2B/7B14B
Context window
8k2k
Benchmarks
HumanEval
32.3%32.3%
MATH
24.3%24.8%
MMLU
64.3%66.3%
GSM8K
46.4%61.3%
Overview
CompanyGoogleQwen
Release dateFeb 21 2024Sep 25 2023
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Qwen-14B leads Gemma 1 on 3 of the 4 benchmarks they both report (HumanEval, MATH, MMLU, GSM8K). Qwen-14B shipped 149 days before Gemma 1, so benchmark comparisons should account for the intervening progress.

Gemma 1 has 2B/7B parameters, while Qwen-14B has 14B. Context windows are 8k (Gemma 1) vs 2k (Qwen-14B).

On HumanEval, both models score 32.3%. On MATH, Qwen-14B leads at 24.8% vs Gemma 1 at 24.3%. On MMLU, Qwen-14B leads at 66.3% vs Gemma 1 at 64.3%. On GSM8K, Qwen-14B leads at 61.3% vs Gemma 1 at 46.4%.